Objectives:To determine the factors that influence the decision maker regarding end-of-life (EOL) care and to disclose how aggressive care in the last week before death, place of death, and quality-of-life (QOL) affected the decisions made. Methods:The subjects were 409 bereaved family members (cause of death included cancer, stroke, heart disease, and pneumonia) that registered with an internet research agency. Decision-making was controlled either by the patient, family, or physician or shared by the patient, family, and physician. Results:The results of a multinomial logistic regression analysis demonstrated that when a family controlled decision-making, they were less likely to report patient-family EOL discussion [odds ratio (OR)=0.52], and that the patient had good communication with the physician (OR=0.77); they were also likely to report that the patient had dementia (OR=1.94). Families who reported physician-controlled decision-making (vs patient controlled) were less likely to report that the patient had good communication with the physician (OR=0.62). Cardiopulmonary resuscitation in the last week and place of death were not associated with the decision maker. EOL QOL was associated with EOL care of the decision maker. When a physician controlled decision-making, patient EOL QOL was at its lowest. Conclusion:To improve patient-controlled decision-making, it is critical that patients, families, and physicians have more communication regarding EOL care.

Objectives: To assess the perception of care and outcomes of end-of-life palliative care by bereaved family members to determine differences in care provided to patients with and without cancer. Methods: This cross-sectional, anonymous survey using a self-reporting questionnaire for bereaved family members was conducted online. Care was assessed using overall satisfaction score and the care evaluation scale (CES) and outcomes were assessed using good death inventory (GDI). Results: The present study included data from 118 patients with cancer and 299 patients without cancer (103, heart failure; 71, stroke; and 125, pneumonia). The overall satisfaction score was not significantly different between patients with and without cancer. Conversely, physical care score in the CES and autonomy score in the GDI were significantly lower in patients without cancer than in patients with cancer (p<0.05). Conclusion: The satisfaction with end-of-life care was comparable between the bereaved family members of patients without cancer and those of patients with cancer. However, results related to some items of CES and GDI suggest that some components of end-of-life care for patients without cancer might require attention. Not only treatment of the underlying disease but also relief of suffering is important to improve end-of-life care.